Despite suggestive experimental and epidemiologic evidence at high-chronic arsenic exposure (>=100 ug/L in drinking water), the effects of low and moderate chronic arsenic exposure on cardiovascular and diabetes risk are unknown. In the US, naturally occurring arsenic in drinking water disproportionately affects rural communities of the Western States, including Native Americans. Objective: To evaluate the association of inorganic arsenic exposure, as determined by the sum of inorganic and methylated arsenic species in urine, and arsenic biotransformation with the risk of cardiovascular disease and diabetes in 4,549 Native Americans who participated in the Strong Heart Study. Hypotheses: (1) Urine arsenic concentrations are positively associated with increased incidence of coronary heart disease, stroke, peripheral arterial disease, and diabetes. (2) The association of arsenic exposure with the study outcomes depends on arsenic biotransformation. (3) The association of arsenic exposure and biotransformation with the study outcomes is modified by smoking status. Preliminary studies: The sum of inorganic and methylated urine arsenic species ranged from 2.1 to 72.8 5 ug/g creatinine (median 7.2 5 ug/g) in 60 Strong Heart Study Participants selected at random. Urine arsenic concentrations were relatively constant over the 10-years of follow-up. Design and setting: Population-based prospective cohort study. From 1989 to 1991, the Strong Heart Study recruited 4,549 Native Americans 45 to 74 years of age residing in Arizona, Oklahoma and North and South Dakota. The study collected sociodemographic, life-style and medical information and conducted clinical examinations of participants during three visits over a 10-year period. Participants were followed for mortality and morbidity through 2001. Urine samples were collected and stored at -70:C in all study participants. Exposure assessment: Arsenic exposure and biotransformation will be assessed in urine samples of all Strong Heart Study participants at baseline. Total urine arsenic and urine arsenic species concentrations will be determined using inductively coupled plasma-mass spectrometry (ICPMS) and high performance liquid chromatography / vapor generation-ICPMS (HPLC/VG-ICPMS), respectively. Statistical analysis: Cox proportional hazard models will be used to assess the prospective association of urine arsenic species concentrations with the incidence of coronary heart disease, stroke, peripheral arterial disease and diabetes. Important statistical issues will be the control of laboratory and biological measurement error, the evaluation of the dose-response through the range of arsenic concentrations, the modeling of urine arsenic species concentrations, and the evaluation of the joint effect of arsenic and smoking. Significance: Tens of millions of people in the US and abroad are exposed to arsenic in drinking water >10 ug/L, the current US standard. The elucidation of the contribution of arsenic exposure to cardiovascular and diabetes development in this high-quality prospective cohort study can inform and potentially impact current environmental public health recommendations. PUBLIC HEALTH RELEVANCE: Tens of millions of people in the US and abroad are exposed to arsenic in drinking water above 10 ug/L, the current US arsenic standard. In the US, naturally occurring arsenic in drinking water disproportionately affects rural communities of the Western States, including Native Americans. The objective of this study is to evaluate the association of arsenic exposure and biotransformation with the risk of cardiovascular disease and diabetes in 4,549 Native Americans who participated in the Strong Heart Study, the largest study of cardiovascular disease in Native Americans. The elucidation of the contribution of arsenic exposure to cardiovascular and diabetes development in this high-quality prospective cohort study can inform and potentially impact current environmental public health recommendations.